Model benchmarks
gemma4:e2b local LLM performance
As of November 2026, gemma4:e2b runs at up to 158.5 tok/s for local inference (best of 37 community benchmark runs across 2 GPUs).
Model size
4.6B
Peak speed
158.5 tok/s
Average speed
151.6 tok/s
Avg PP
9602.8 tok/s
Min memory
369 MB
Max context
131,072 tokens
Avg output / run
8,705 tokens
Avg runtime / run
2m 6s
Avg quality
68.2
Benchmark runs
37
GPUs tested
2
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 37 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 61.2 | 68.2 | 73.8 |
| Agent Workflow | 42.3 | 71.5 | 89.1 |
| Code Generation | 41.6 | 56.7 | 67.4 |
| Role Play & Narrative | 62.9 | 75.8 | 87.9 |
| Research & Analysis | 59.4 | 68.6 | 76.5 |
Performance by hardware and tool
Every hardware/tool/quantization combination gemma4:e2b has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA RTX 3500 Ada Generation Laptop GPU | Ollama | Q4_K_M | 158.5 tok/s | 153.3 tok/s | 369 MB | 131,072 tokens | 68.2 | 36 |
| NVIDIA GeForce RTX 4060 Laptop GPU | Ollama | Q4_K_M | 90.7 tok/s | 90.7 tok/s | 1.6 GB | 8,192 tokens | 66.3 | 1 |
Benchmark runs
All 37 gemma4:e2b runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is gemma4:e2b good for agentic (tool-using) tasks?
- In our benchmarks, gemma4:e2b scores 71.5/100 for agentic workflows. It runs at about 151.6 tok/s, so if you want more speed, unsloth/Qwen3.6-35B-A3B-GGUF:UD-Q4_K_M is faster (~246.8 tok/s) and still scores well for agentic workflows (80.6/100).
- How fast is gemma4:e2b for local inference?
- Across 37 community benchmark runs, gemma4:e2b reaches up to 158.5 tok/s and averages 151.6 tok/s, with the fastest results on NVIDIA RTX 3500 Ada Generation Laptop GPU.
- How much memory does gemma4:e2b need?
- The leanest observed configuration used about 369 MB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run gemma4:e2b?
- Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.